Video to Text Converter

What people turn videos into text for

Meetings

Minutes and action items from Zoom, Teams and Meet recordings — see meeting transcription.

Lectures

Searchable notes from recorded classes — see lecture transcription.

Content

Blog posts, show notes and quotes from videos you've published.

Accessibility

Text and captions for viewers who can't hear the audio.

Priced for long recordings

Transcription costs 1 credit per minute — 60 credits (6¢) for an hour, 120 credits (12¢) for two. Adding a translated transcript costs nothing extra.

How it works

1. Drop in a video. Your browser reads it and pulls out only the audio — the video file never leaves your device.
2. The audio is processed in 30-second chunks: speech is transcribed and the spoken language is detected automatically.
3. Download the results. Audio and text on our servers are deleted after delivery, within 30 minutes at most.

What you get

Transcript

Plain text (.txt) of what was said

Timestamped transcript

Every line prefixed with its time, e.g. [12:04]

Subtitles

SRT and VTT, in the language you select (select the spoken language for untranslated subtitles)

Supported files

• Video: MP4, WebM, MOV, MKV, M4V — whatever your browser can read.
• Length: up to 2 hours per file.
• Languages: speech in 21 languages is recognised automatically; translation and voices cover the same 21.

Privacy

Your video is never uploaded — only its audio is, over HTTPS in small chunks. Audio, transcripts and voice files are deleted after delivery, within 30 minutes at most, and are never used to train models. Your account keeps only the file name, length, languages and credits used. Details in the privacy policy.

Good to know

• No speaker labels — transcripts don't say who is speaking.
• Files only: no links from YouTube or social apps, and no live audio.
• Subtitles come as SRT/VTT files rather than burned into the picture.

Picking the right download for the job

Transcript mode gives four files, and each suits a different next step:

Quoting someone in an article or report

Timestamped transcript: find the line, cite the time, jump back to check the wording

Turning a talk into a blog post or handout

Plain transcript: fewer timestamps to strip out

Captions for your player, course platform or YouTube

SRT or VTT

Reading a video in a language you don't speak

Pick a target language; the transcript download and subtitles follow it, and the result screen shows the original alongside

Pick a target language and one run gives both texts: Download transcript is the translation and Download original transcript is the wording as spoken. Subtitles come in one language per run, so captions in the spoken language plus a translated set means a second run; on a half-hour video that costs another 30 credits (3¢). Transcription or translation? explains how the two steps fit together.

Three things to check before uploading

• More than one audio track: camera files and some screen recordings carry several, and the one with speech isn't always the first. Videos with more than one audio track shows how to check and pick.
• Long music-only stretches: speech models can write words that nobody said during intros, outros and music beds. Trimming a two-minute title sequence is quicker than deleting phantom lines afterwards.
• Recordings longer than two hours: split at a natural pause, such as a break in a livestream or between conference sessions, rather than at an arbitrary time.
• Speakers far from the microphone: in a room recording, people far from the camera come out with more errors. If a lapel or desk microphone also recorded the session, transcribe that file instead of the camera's audio.

Cleaning up a machine transcript

1. Search for names, product terms and numbers first; these are where recognition errors cluster.
2. Use the timestamps to jump to doubtful passages instead of re-listening to the whole video.
3. Add speaker names by hand. Transcripts don't label speakers, so a Q&A or interview needs names inserted where the voice changes.
4. Decide how verbatim you need it. Speech models of this kind tend to smooth over ums and false starts; if your work needs true verbatim, add them by ear.
5. Fix sentence breaks last, once the words are right.

The proofreading method for AI transcripts covers how to do this without rereading every line, and from transcript to article picks up where the clean transcript ends.

Worked examples

Where the text goes next

A transcript is rarely the final product. The common next steps each favour a different download:

• Searching a back catalogue: keep the plain transcripts in one folder next to the videos, named the same way, and ordinary desktop search finds the video where something was said.
• Cutting clips: scan the timestamped transcript for strong lines, note the times, and cut in your editor. Finding clips with a transcript shows the method.
• Publishing alongside the video: a cleaned transcript under an embedded video helps readers who prefer text and anyone who can't play sound. Formatting a raw transcript covers paragraphs, headings and speaker names.
• Sharing with colleagues who read another language: send the translated transcript and keep the original for checking quotes.

What a video transcript won't contain

• Who is speaking: there are no speaker labels.
• Words that appear only in the picture, like slide text or a caption already burned into the footage. Only speech is transcribed.
• Sound descriptions such as laughter or music cues.
• A summary: mydubly returns the full text and doesn't summarise it.

For the spoken language, see the pages for Hindi, where Hindi and English often mix in one sentence, Japanese, which is written without spaces between words, and Arabic. MP4 files covers the most common container, and the video transcription guide walks through a full run.